[PATCHv2 0/4] IB cache port state, use it when resolve address

[PATCHv2 0/4] IB cache port state, use it when resolve address

[Jinpu Wang]

Hi Folks,

This is my forth attempt to fix rdma_revolve_addr may resolve to first
port even it's down[1].

As Jason and Hefty suggested, I now cache port_state in ib_core, and
update cma_resolve_ib_dev and cma_bind_loopback to use it.

0001-IB-core-add-port-state-cache.patch
0002-IB-core-export-ib_get_cached_port_state.patch
0003-cma-resolve-to-first-active-ib-port.patch
0004-cma-use-cached-port-state-when-bind-loopback.patch

[1]http://marc.info/?l=linux-rdma&m=148102306632521&w=2

Changes from v1:
- Add Hefty's Acked-by.
- rebase on Doug's rdma/k.o/for-4.10 (commit
5cc8fabc5e4c588c75a5ec21423e7c3425f69f48)
- fix a mem leak in ib_cache_setup_one in error case in first patch.

-- 
Jinpu Wang
Linux Kernel Developer

ProfitBricks GmbH
Greifswalder Str. 207
D - 10405 Berlin

Tel:       +49 30 577 008  042
Fax:      +49 30 577 008 299
Email:    jinpu.wang-EIkl63zCoXaH+58JC4qpiA@public.gmane.org
URL:      https://www.profitbricks.de

Sitz der Gesellschaft: Berlin
Registergericht: Amtsgericht Charlottenburg, HRB 125506 B
Geschäftsführer: Achim Weiss
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 0/4] IB cache port state, use it when resolve address

[Jinpu Wang]

On Mon, Jan 2, 2017 at 1:15 PM, Jinpu Wang <jinpu.wang-EIkl63zCoXaH+58JC4qpiA@public.gmane.org> wrote:
> Hi Folks,
>
> This is my forth attempt to fix rdma_revolve_addr may resolve to first
> port even it's down[1].
>
> As Jason and Hefty suggested, I now cache port_state in ib_core, and
> update cma_resolve_ib_dev and cma_bind_loopback to use it.
>
> 0001-IB-core-add-port-state-cache.patch
> 0002-IB-core-export-ib_get_cached_port_state.patch
> 0003-cma-resolve-to-first-active-ib-port.patch
> 0004-cma-use-cached-port-state-when-bind-loopback.patch
>
> [1]http://marc.info/?l=linux-rdma&m=148102306632521&w=2
>
> Changes from v1:
> - Add Hefty's Acked-by.
> - rebase on Doug's rdma/k.o/for-4.10 (commit
> 5cc8fabc5e4c588c75a5ec21423e7c3425f69f48)
> - fix a mem leak in ib_cache_setup_one in error case in first patch.
>
> --
Ping Doug?

-- 
Jinpu Wang
Linux Kernel Developer

ProfitBricks GmbH
Greifswalder Str. 207
D - 10405 Berlin

Tel:       +49 30 577 008  042
Fax:      +49 30 577 008 299
Email:    jinpu.wang-EIkl63zCoXaH+58JC4qpiA@public.gmane.org
URL:      https://www.profitbricks.de

Sitz der Gesellschaft: Berlin
Registergericht: Amtsgericht Charlottenburg, HRB 125506 B
Geschäftsführer: Achim Weiss
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 0/4] IB cache port state, use it when resolve address

[Doug Ledford]

[-- Attachment #1: Type: text/plain, Size: 2014 bytes --]

On Wed, 2017-01-11 at 13:26 +0100, Jinpu Wang wrote:
> On Mon, Jan 2, 2017 at 1:15 PM, Jinpu Wang <jinpu.wang@profitbricks.c
> om> wrote:
> > 
> > Hi Folks,
> > 
> > This is my forth attempt to fix rdma_revolve_addr may resolve to
> > first
> > port even it's down[1].
> > 
> > As Jason and Hefty suggested, I now cache port_state in ib_core,
> > and
> > update cma_resolve_ib_dev and cma_bind_loopback to use it.
> > 
> > 0001-IB-core-add-port-state-cache.patch
> > 0002-IB-core-export-ib_get_cached_port_state.patch
> > 0003-cma-resolve-to-first-active-ib-port.patch
> > 0004-cma-use-cached-port-state-when-bind-loopback.patch
> > 
> > [1]http://marc.info/?l=linux-rdma&m=148102306632521&w=2
> > 
> > Changes from v1:
> > - Add Hefty's Acked-by.
> > - rebase on Doug's rdma/k.o/for-4.10 (commit
> > 5cc8fabc5e4c588c75a5ec21423e7c3425f69f48)
> > - fix a mem leak in ib_cache_setup_one in error case in first
> > patch.
> > 
> > --
> Ping Doug?

Sorry, I didn't see this before I wrote my last email.  You've already
captured Sean's acks, that's good.  You modified the first patch in a
way that negates the comment request I made, so that's fine.  But the
patches are still coming through mangled.  If you look at them in
patchworks, you can see telltale signs, such as seeing part of the
patch in the patchwork comments and then the full patch later on in the
patch section.  When I download those patches, they are mangled.  You
can *not* both send the patch inline and as an attachment.  The
prefered way to send patches, if your mailer is messing with your
submissions, is to use "git send-email" and let it send the patches
directly for you.

However, this time I have fixed the patches up and applied them, but
please don't send mangled patches like this in the future.

-- 
Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
    GPG KeyID: B826A3330E572FDD
   
Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCHv2 0/4] IB cache port state, use it when resolve address

[Jinpu Wang]

On Fri, Jan 13, 2017 at 5:02 AM, Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> On Wed, 2017-01-11 at 13:26 +0100, Jinpu Wang wrote:
>> On Mon, Jan 2, 2017 at 1:15 PM, Jinpu Wang <jinpu.wang@profitbricks.c
>> om> wrote:
>> >
>> > Hi Folks,
>> >
>> > This is my forth attempt to fix rdma_revolve_addr may resolve to
>> > first
>> > port even it's down[1].
>> >
>> > As Jason and Hefty suggested, I now cache port_state in ib_core,
>> > and
>> > update cma_resolve_ib_dev and cma_bind_loopback to use it.
>> >
>> > 0001-IB-core-add-port-state-cache.patch
>> > 0002-IB-core-export-ib_get_cached_port_state.patch
>> > 0003-cma-resolve-to-first-active-ib-port.patch
>> > 0004-cma-use-cached-port-state-when-bind-loopback.patch
>> >
>> > [1]http://marc.info/?l=linux-rdma&m=148102306632521&w=2
>> >
>> > Changes from v1:
>> > - Add Hefty's Acked-by.
>> > - rebase on Doug's rdma/k.o/for-4.10 (commit
>> > 5cc8fabc5e4c588c75a5ec21423e7c3425f69f48)
>> > - fix a mem leak in ib_cache_setup_one in error case in first
>> > patch.
>> >
>> > --
>> Ping Doug?
>
> Sorry, I didn't see this before I wrote my last email.  You've already
> captured Sean's acks, that's good.  You modified the first patch in a
> way that negates the comment request I made, so that's fine.  But the
> patches are still coming through mangled.  If you look at them in
> patchworks, you can see telltale signs, such as seeing part of the
> patch in the patchwork comments and then the full patch later on in the
> patch section.  When I download those patches, they are mangled.  You
> can *not* both send the patch inline and as an attachment.  The
> prefered way to send patches, if your mailer is messing with your
> submissions, is to use "git send-email" and let it send the patches
> directly for you.
>
> However, this time I have fixed the patches up and applied them, but
> please don't send mangled patches like this in the future.
>
Thanks Doug for taking care of the format.
I will definitely will configure and my email client properly when send patch
to mainline.
-- 
Jinpu Wang
Linux Kernel Developer

ProfitBricks GmbH
Greifswalder Str. 207
D - 10405 Berlin

Tel:       +49 30 577 008  042
Fax:      +49 30 577 008 299
Email:    jinpu.wang-EIkl63zCoXaH+58JC4qpiA@public.gmane.org
URL:      https://www.profitbricks.de

Sitz der Gesellschaft: Berlin
Registergericht: Amtsgericht Charlottenburg, HRB 125506 B
Geschäftsführer: Achim Weiss
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html